Most ammunition is produced long before its ultimate consumption and stored in a series of different depots for a considerably long period of time. During storage, the quality of the ammunition stockpile deteriorates proportionally to the conditions of depots. We view different conditions associated with a series of depots as step-stress. A random effects logistic regression model is employed to predict the quality of ammunition stockpile in terms of the routing information such as a series of location and duration of storage of ammunition lots. The resultant prediction model can be used to determine the appropriate time for reorder or renovation of ammunition before the quality reaches substandard. An example is given to illustrate the implementation procedure of the prediction model suggested in this paper
aq/aq cc:9116 08/06/98
Canon EOS 5D Mark II
Naval Postgraduate School (U.S.). Dept. of Operations Research.